metric

`recall_10`

better: higher
kind: quality: measured against a reference or by a judge
decides: no goal
weighed by: no goal
recorded on: scifact_200

01what it measures, and how

Share of the query's relevant documents found in the first 10.

A project explains its own metric at length with METRIC_DESCRIPTIONS = {"recall_10": {"summary": "…", "details": "…"}}.

02on scifact_200

scored against silverscored by a judgewrote the reference it is scored againstscored against a label set, not a reference setunmarked rows are scored against gold

arm	mean recall_10	items	scored against
sf_glm_s_e5first_n200_v1 · pair	0.860	200	gold/scifact_200 gold
sf_qwen_s_e5first_n200_v1 · pair	0.856	200	gold/scifact_200 gold
sf_e5_base_n200_v1	0.831	200	gold/scifact_200 gold
sf_bge_small_n200_v1	0.830	200	gold/scifact_200 gold
sf_glm_s_bm25first_n200_v1 · pair	0.812	200	gold/scifact_200 gold
sf_glm_m_n200_v1	0.807	200	gold/scifact_200 gold
sf_glm_s_n200_v1	0.807	200	gold/scifact_200 gold
sf_deepseek_s_n200_v1	0.802	200	gold/scifact_200 gold
sf_gemma_m_n200_v1	0.797	200	gold/scifact_200 gold
sf_qwen_s_n200_v1 · repeats	0.797	200	gold/scifact_200 gold
sf_llama_l_n200_v1	0.795	200	gold/scifact_200 gold
sf_qwen_s_n200_v1	0.795	200	gold/scifact_200 gold
sf_gemma_l_n200_v1	0.792	200	gold/scifact_200 gold
sf_llama_l_n200_v1 · repeats	0.792	200	gold/scifact_200 gold
sf_qwen_s_bm25first_n200_v1 · pair	0.790	200	gold/scifact_200 gold
sf_deepseek_m_n200_v1	0.789	200	gold/scifact_200 gold
sf_llama_s_n200_v1	0.785	200	gold/scifact_200 gold
sf_llama_m_n200_v1	0.779	200	gold/scifact_200 gold
sf_gemma_s_n200_v1	0.776	200	gold/scifact_200 gold
sf_llama_s_n200_v1 · repeats	0.775	200	gold/scifact_200 gold
sf_mpnet_n200_v1	0.771	200	gold/scifact_200 gold
sf_minilm_n200_v1	0.770	200	gold/scifact_200 gold
sf_mistral_s_n200_v1	0.767	200	gold/scifact_200 gold
sf_bm25_n200_v1	0.763	200	gold/scifact_200 gold
sf_random_n200_v1	0.001	200	gold/scifact_200 gold
sf_first_k_n200_v1	0.000	200	gold/scifact_200 gold